PROJECT SUMMARY Biomarkers for the early detection of pancreatic cancer are urgently needed. However, individual molecules with the sensitivity and specificity needed for population-based screening have not been discovered. CA-19-9 has been studied extensively and yet has failed to demonstrate the predictive value necessary for early detection and diagnosis. Although many platforms, including proteomic, genomic and transcriptomic approaches have been utilized and biomarker candidates identified, no one platform or molecule has been successfully validated in large blinded population screens. We have taken a functional genomic pathways approach to biomarker discovery targeting the earliest genomic intervals altered in pancreatic and other smoking related cancers, focusing on biomarkers involved in critical cancer relevant cellular pathways. We have identified a ?migration signature? and biomarker panel that has gone through two blinded validations including the NCI-EDRN pancreatic cancer reference set of early stage disease and benign and healthy controls. Results indicate that our panel of biomarkers improves the performance of CA19-9 to detect asymptomatic early stage pancreatic cancer yielding significant results across validations platforms. We have developed plasma miRNA biomarker panels for pancreatic cancer and validated these in multiple trials. The combination of our biomarkers demonstrates high sensitivity and specificity to detect early stage pancreatic cancer. We will further refine this early stage biomarker panel and a novel risk score approach to stratify the general population and at risk population for screening. Refined panels will be validated in large retrospective and prospective cohorts for early detection of pancreatic cancer according to ProBE design. In order that we identify early stage disease prior to metastasis, amenable to curative intervention, we will utilize a mouse- human approach to develop integrated biomarkers profiles from plasma and plasma exosomes to detect late precursor stage PanINs prior to development of advanced cancer with subsequent validation for a prediagnostic biomarker panel. We will also develop a novel two tiered screening strategy with validated biomarkers and novel imaging tools to change the clinical management plan and significantly improve survival of one of the most deadly cancers.